MingyuanLuo

在macro.tex末尾添加 ```tex \setlength{\bibsep}{2.8pt} ```

将macro.tex中的以下代码(line 47) ```tex \usepackage{titlesec} \usepackage{titletoc} \usepackage{tocloft} \titlecontents{section}[1em]{\zihao{-4}}{\contentslabel{1em}}{\hspace{-1em}}{\titlerule*[0.5pc]{$.$}\contentspage} \titlecontents{subsection}[3em]{\zihao{-4}}{\contentslabel{2em}}{\hspace{-2em}}{\titlerule*[0.5pc]{$.$}\contentspage} \titlecontents{subsubsection}[5em]{\zihao{-4}}{\contentslabel{3em}}{\hspace{-3em}}{\titlerule*[0.5pc]{$.$}\contentspage} \renewcommand{\cfttoctitlefont}{\zihao{2}\heiti} \renewcommand{\contentsname}{\hfill 目~录\hfill} \renewcommand{\cftaftertoctitle}{\hfill} ``` 修改为： ```tex \usepackage{titlesec} \usepackage{titletoc} \usepackage{tocloft} \titlecontents{section}[1em]{\zihao{-4}\hei}{\contentslabel{1em}}{\hspace{-1em}}{\titlerule*[0.5pc]{$.$}\contentspage\vspace{2.8pt}} \titlecontents{subsection}[3em]{\zihao{-4}\hei}{\contentslabel{2em}}{\hspace{-2em}}{\titlerule*[0.5pc]{$.$}\contentspage\vspace{2.8pt}} \titlecontents{subsubsection}[5em]{\zihao{-4}\hei}{\contentslabel{3em}}{\hspace{-3em}}{\titlerule*[0.5pc]{$.$}\contentspage\vspace{2.8pt}} \renewcommand{\cfttoctitlefont}{\zihao{2}\hei} \renewcommand{\contentsname}{\hfill 目~录\hfill} \renewcommand{\cftaftertoctitle}{\hfill} ```

You are right about the performance, I have only tested it with one NPC yet. I will try to find test cases with more NPC to see if the performance...

Thanks for your reply ! I will try to find a better solution.

Thanks! This is a very nice case to understand why you say that no replan in last section is a design choice.

However, I think it's better for ego to still have some lightly ability to avoid, but not too strong ability. For your example, I think the second bus should preserve...

@hatem-darweesh Thank you! There are two more details: 1. TrajectoryEvaluator::blockTrajectoryOutOfRoad() is invoked in TrajectoryEvaluator::doOneStep() 2. I add a RoadNetwork m_Map in class TrajectoryEvaluator, which is used in **blockTrajectoryOutOfRoad**. And the...

Thanks @hatem-darweesh, I will try to do some optimization. if the results are good, I will try to send a pull request for this solution.